The present invention relates to methods for determination of the presence of Lipopolysaccharide binding protein (LBP) in body fluid samples including blood samples.
Lipopolysaccharide (LPS) is a common component of the outer membrane of Gram-negative bacteria and is responsible for many of the pathologic effects associated with gram-negative bacterial infection and endotoxemia. Because of the association between bacterial infection and sepsis, attempts have been made to correlate serum/plasma levels of endotoxin with disease. Typically, endotoxin levels have been measured using the Limulus amebocyte lysate (LAL) assay, in which endotoxin initiates a coagulation cascade that can be measured physically, turbidimetrically, or spectrophotometrically, Despite these attempts, however, no reliable correlations between endotoxin levels and sepsis severity or outcome have been identified. This is most likely due to the fact that (i) endotoxin levels in septic patients are very low (&gt;10 pg/L), several serum proteins interfere with the proteolytic LAL cascade, (iii) endotoxin, once in contact with blood, can be "detoxified" by interaction with a variety of blood components, including high-density lipoprotein (HDL) and low-density lipoprotein (LDL) and (iv) endotoxin from different gram-negative organisms varies in its ability to trigger the LAL cascade. Thus, the absolute levels of endotoxin in a patient sample may not correspond to the actual concentrations of bioactive endotoxin present in vivo.
Two proteins have been identified in humans and other animals that bind LPS with high affinity. These two proteins, Lipopolysaccharide binding protein (LBP), and bactericidal/permeability increasing protein (BPI) have roughly the same molecular weight and share 45% amino acid homology, yet exhibit distinct physiological differences. LBP is a 60 kD glycoprotein synthesized in the liver, while BPI is found in the azurophilic granules of neutrophils. LBP is found in the serum of normal humans at levels of 5-10 .mu.g/mL but can reach levels of 50-100 .mu.g/mL in septic patients. Schumann et al., Science, 249:1429 (1990) disclose the amino acid sequences and encoding cDNA of both human and rabbit LBP. Like BPI, LBP has a binding site for lipid A and binds to the LPS from rough (R-) and smooth (S-) form bacteria. Unlike BPI, LBP does not possess significant bactericidal activity. BPI has been observed to neutralize and inhibit the production of TNF resulting from interaction of LBP with LPS and CD14 on monocytes and macrophages. Marra et al., J. Immunol. 148:532 (1992), Weiss et al., J. Clin. Invest. 90:1122 (1992). In contrast, LBP is observed to enhance LPS-induced TNF production. Wright et al., Science, 249:1131 (1990). Thus, in contrast to BPI, LBP has been recognized as an immunostimulatory molecule. See, e.g., Seilhamer, PCT International Application WO 93/06228 which discloses a variant form of LBP which it terms LBP-.beta.. Also of interest to the present invention are Ulevitch, PCT International Application WO 91/01639 which discloses, among other things, anti-LBP antibodies as an anti-sepsis therapeutic agent and U.S. Pat. No. 5,245,013 which relates to LBP and discloses antibodies which immunoreact with a polypeptide having homology to LBP.
LBP has been characterized in the art as an "acute phase protein", that is one of several plasma proteins such as C-reactive protein or serum amyloid A that increase in concentration in response to infectious and non-infectious tissue destructive processes. As such, it would be anticipated that LBP levels would be elevated in samples from patients suffering from a number of autoimmune diseases such as rheumatoid arthritis and lupus erythematosus. Of interest to the present invention are disclosures related to the assaying of BPI activity in subjects. von der Mohien et al., Abstract, 13th International Symposium on Intensive Care and Emergency Medicine, Brussels (March 1993) discloses the results of assays for serum levels of BPI in patients with gram-negative sepsis and healthy subjects. The abstract disclosed that no BPI was detectable under the conditions of the assay in the serum of healthy subjects while circulating BPI was detected in all septic patients. Also of interest is the disclosure of co-owned and copending U.S. patent application Ser. No. 08/175,276 filed Dec. 29, 1993 which is a continuation-in-part of application Ser. No. 08/125,677 filed Sep. 22, 1993 the disclosures of which are hereby incorporated by reference. Those patent applications disclose that levels of BPI in blood plasma samples correlate with the presence or absence of sepsis while levels of BPI in blood serum samples do not. The patent applications teach that levels of BPI present in serum are not representative of endogenous levels of BPI in circulating blood while levels of BPI in plasma are.
Also of interest to the present invention are the disclosures of Leturcq et al., Keystone Tahoe Endotoxin Conference, Mar. 1-7, 1992 (Abstract) in which the generation of monoclonal antibodies to human LBP is reported. Also reported is the screening of normal human serum samples for the presence of LBP. LBP levels for normal serum samples were reported to range from 1 .mu.g/mL to 24 .mu.g/mL with an average of 7 .mu.g/mL. Further of interest is the disclosure of Richard Ulevitch at the American Society for Microbiology General Meeting in Atlanta, Ga. May 16-21 (1993) at which data was presented on LBP and soluble CD14 levels in the serum of septic and healthy individuals. The average soluble CD14 and LBP concentrations in the serum of healthy adults were 1 .mu.g/mL and 7 .mu.g/mL respectively. The average soluble CD14 and LBP concentrations in the serum of septic patients were reported to be 2 .mu.g/mL and 55 .mu.g/mL respectively.
Geller et al., Arch. Surg., 128:22-28 (1993) disclose experiments in which the induction of LBP mRNA was studied in three models known to induce acute phase responses: (1) LPS injection; (2) Corynebacterium parvum injection; and (3) turpentine injection. The publication reports that LBP mRNA is induced during hepatic inflammation and suggest that LBP is an acute-phase protein important in regulating the in vivo response to endotoxin.
Gallay et al., Infect. Immun., 61:378-383 (1993) disclose that an acute phase response in mice injected with silver nitrate induced LBP synthesis, and that LBP levels increase approximately 10-fold over normal levels after an acute-phase response.
There exists a desire in the art for methods for determining the exposure of subjects to endotoxin and for distinguishing the effects of exposure to endotoxin from other acute phase physiologic responses. Also desired are methods for diagnosing the presence or severity of gram-negative sepsis in a subject and for predicting the prognosis of a subject suffering from sepsis.